OpenAGI is a package for AI agent creation designed to connect large language models with domain-specific tools and workflows in the AIOS (AI Operating System) ecosystem. It provides a structured Python framework, pyopenagi, for defining agents as modular units that encapsulate execution logic, configuration, and dependency metadata. Agents are organized in a well-defined folder structure that includes code (agent.py), configuration (config.json), and extra requirements (meta_requirements.txt), which makes them easy to package, share, and reuse. The project includes tooling for registering agents with AIOS by uploading them via a command-line interface, enforcing a consistent naming scheme that matches the local folder layout. A companion tooling layer lets agents call external tools described in the tools.md documentation, enabling them to orchestrate APIs, retrieval pipelines, and other utilities in response to LLM decisions.
Features
- Python-based agent framework (pyopenagi) tailored for AIOS integration
- Prescribed agent folder structure with execution code, configuration, and dependency metadata
- CLI workflow for packaging and uploading agents to a shared AIOS environment
- Tooling specification system so agents can invoke external tools and services
- Support for local development via editable installs and test suites
- Acts as a reference and legacy SDK for AIOS agent creation alongside the newer Cerebrum SDK